Austin
Harrison

The Attribution Skeptic Who Became the Standard

There's a data point inside Northbeam's platform that Austin Harrison likes to surface in interviews: most brands, when they look at their marketing data carefully for the first time, discover they've been misattributing a significant chunk of their revenue. Not by a rounding error. By a lot. Harrison built a company around fixing that — not by adding more dashboards, but by rethinking the underlying data collection from scratch.

Northbeam's approach rests on three columns: first-party data captured directly from brand storefronts, machine learning models that process it at scale, and custom attribution frameworks built for the reality of multi-channel commerce. The company calls the result Universal Attribution. In practice, it means a brand selling luggage can finally see whether the Instagram Reel someone watched on a Tuesday actually influenced the purchase they made on Friday — rather than crediting the final click that happened to land the sale.

Harrison's entry point into this problem was not a typical one. He did not come out of performance marketing or growth hacking. He came out of audience measurement — specifically, helping build the random sampling methodology at NetRatings, the firm that eventually had 60% of itself acquired by Nielsen. That work, unglamorous as it sounds, drilled into him an obsession with measurement integrity that later defined everything Northbeam built.

Cartoons, Transformers, and a Side Career That Never Stopped

In 1999, Harrison was running MediaTrip, an online animation channel that Entertainment Weekly named one of its top Internet Phenomena of the year. The channel was nominated for an HBO comedy award. This was pre-YouTube, pre-streaming, before anyone had a coherent business model for internet video — and Harrison was already figuring out how audiences found content online.

By 2005 he had co-founded Harrison Brooks LLC, an animation development and production company. Its clients included Hasbro — specifically for the development of a Transformers animated TV series — along with Disney, NBC Universal, and projects tied to Kelsey Grammer. The company also produced award-winning graphic novels and collaborated with the rapper Common on creative projects. This is not the typical resume item that precedes a Series A pitch, but Harrison has never seemed especially interested in doing what's typical.

The animation chapter never fully closed. In March 2025, Harrison was listed as a collaborator on Churp — a new cartoon and merchandise subscription service launched by Titmouse, the studio behind Big Mouth, at their SXSW 25th anniversary panel. Each month, Churp subscribers get a mystery box of bespoke apparel and collectibles, plus a secret code unlocking exclusive hand-crafted cartoons. It's the kind of side project that, from a CEO, raises eyebrows. From Harrison, it reads like continuity.

From Dentsu to Startup: The Long Route to Northbeam

After Harrison Brooks, Harrison moved into the corporate end of digital media. From 2008 to 2011, he ran the digital media team at Dentsu America, one of the world's largest advertising networks. This gave him a seat inside the machine — watching how large agencies allocated spend, analyzed performance, and often operated with enormous confidence on incomplete data.

The years that followed included a series of operator and advisory roles at companies across sectors: Luminar Technologies, Sweetfin (the poke bowl chain), Structure Capital, the assessment startup Imbellus, and Noble Transmission. This pattern — varied roles, different industries, no one obvious through-line — suggests someone accumulating vantage points rather than climbing a predetermined ladder.

Northbeam came out of a specific convergence. Harrison had been thinking about machine learning's potential in attribution when he met Dan Huang, a Stanford-trained engineer who had spent time at Twitter building their trending algorithm and then at MTailor applying AI to fashion. Huang had watched a D2C brand waste serious money on campaigns that couldn't be measured properly. The two of them decided to fix it.

They founded Northbeam in 2019. The first version of the platform grew through word-of-mouth — no paid acquisition, no splashy launch. The product itself was doing the selling.

37x and the Series A

Between July 2021 and July 2022, Northbeam's revenue grew 37x. The number is striking enough that it warrants context: the company had been disciplined about product quality over growth, so when growth came, it came fast. Brands like The Ridge, Jones Road Beauty, TRUFF, Vessi, PetMeds, and HexClad were among early commercial partners.

In August 2022, Silversmith Capital Partners led a $15 million Series A. Harrison's statement at the time was characteristically direct: the macroeconomic pressure brands were feeling in 2022 made accurate attribution more important, not less. When budgets tighten, the question of which ad dollars are actually working becomes urgent rather than aspirational.

The platform kept expanding. In 2024, Northbeam launched Apex, a direct integration with Meta designed to feed cleaner first-party data back to Meta's own optimization systems — rather than trying to out-engineer Meta's algorithm, the idea was to give it better inputs. Harrison has been consistent on this point: platforms like Meta and Google have superior data science capabilities. The job is not to compete with that. The job is to make sure they're working with accurate information.

The Clicks + Deterministic Views Moment

October 2025 was a significant month. Northbeam launched what it called an industry-first attribution model: Clicks + Deterministic Views. The product was developed in direct partnership with Meta, TikTok, Snapchat, Pinterest, Axon, MNTN, and Vibe — meaning the platform integrations were not scraped or approximated, but built with the cooperation of the platforms themselves.

The model addresses a gap that had been widening as video advertising grew: when someone watches a Reel or a TikTok but doesn't click, what happens next? Traditional attribution either ignores the view entirely or makes statistical guesses. Northbeam's approach ties first-party transaction data directly to in-app views and clicks across platforms, bypassing the statistical modeling that most measurement tools rely on.

Early results from brand testing showed a 283% lift in attributed transactions for a luggage brand, 175% lift in attributed revenue for a CPG company, and over 550% lift in attributed transactions for a cosmetics company. These are not minor optimizations — they're signals that significant revenue was previously invisible to brands running view-heavy ad campaigns.

Alongside the launch, Northbeam announced a $15 million growth investment led by HighPost Capital — the private equity firm co-founded by Mark Bezos and David Moross — alongside returning investor Silversmith Capital Partners, with participation from Lorimer Ventures and Dynamism Capital. Total funding raised reached $40 million. Harrison called it a defining moment for the company.

What He's Actually Building

Harrison's stated goal is to help every company grow profitably. The practical version of that aspiration is a platform where the question "which of my ads are working?" has a defensible, real-time answer — regardless of whether the user clicked, watched, or simply encountered the ad on their way to something else.

In April 2026, Northbeam launched Northbeam Incrementality, a new advertising measurement standard focused on isolating the true causal impact of ad spend — separating what advertising actually caused from what would have happened anyway. It's the same rigorous thinking that started with random sampling at NetRatings, now applied to the question every brand CMO needs answered before they open next year's budget.

Harrison runs a team of roughly 100 people across engineering, product, and go-to-market. His CTO is Dan Huang, his co-founder. The company is headquartered in San Francisco. He still tweets about marketing data. And somewhere in his schedule, there's apparently still time to develop cartoon subscription services with animation studios.